chassahowitzka national wildlife refuge status and trends

CHASSAHOWITZKA
NATIONAL WILDLIFE REFUGE
STATUS AND TRENDS
A Report
Submitted to:
Ellen M. Porter
Air Quality Branch
U.S. Fish and Wildlife Service
Denver, Colorado
Submitted by :
L. Kellie Dixon
Ernest D. Estevez
Mote Marine Laboratory
1600 Ken Thompson Parkway
Sarasota, Florida 34236
July 10, 1998
Mote Marine Laboratory Technical Report Number 579
This document is printed on recycled paper.
Suggested reference:
Dixon LK, Estevez ED. 1998.
Chassahowitzka National Wildlife Refuge status and trends.
U.S. Fish and Wildlife Service, Air Quality Branch. Mote Marine
Laboratory Technical Report no. 579. 37 p. and appendices.
Available from: Mote Marine Laboratory Library.
ACKNOWLEDGMENTS
Research sponsored by the U.S. Fish & Wildlife Service - Air Quality Branch, and Mote Marine
Laboratory. Ellen Porter, Cam Shaw, Joyce Kleen, Rolf Olson, Bob Quarles, Jay Sprinkel, Jon
Perry, Patricia Minotti, Ari Nissanka, Amy Rhues, and Dimitrious Papadimitriou provided
valuable assistance. Kevin Summers and the GIS staff of EPA Gulf Breeze Laboratory aided in
study design and station mapping.
TABLE OF CONTENTS
Page No.
ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i
TABLE OF CONTENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii
LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii
LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..v
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..l
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..l
MATERIALS AND METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
RESULTS AND DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
StudyArea................................................
Weather.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SAVSampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Water Chemistry - Physical Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Water Chemistry - Nutrients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Phytoplankton . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Biomass.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Sediments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
11
ll
12
13
17
22
28
29
31
33
35
LITERATURE CITED . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
APPENDICES
Appendix A: SAV Quadrat Data
Station and Species Summary
Appendix B: In Situ Water Quality Data
Appendix C: Water Quality Data
Phytoplankton Data
Trophic State Indices
Appendix D: Biomass Data
Appendix E: Sediment Nutrient Data
Sediment Grain Size Data
ii
LIST OF FIGURES
Page No.
Figure 1.
Chassahowitzka National Wildlife Refuge, Chassahowitzka Florida. Thirty coastal
stations, secondary spring vents sampled (CH#l, CRAB, POTTR),
meteorological station (MET), and water level station (WL). . . . . . . . . . . . . . . . . 6
Figure2.
Chassahowitzka and Homosassa River gradient stations and the tidal creek stations
at Seven Cabbage Cutoff (7CABG), Rose Creek (ROSE), Mason Creek
(MASON), upper Mason Creek (UMAS), and Petty Creek (PETTY) . . . . . . . . 10
Figure 3.
Relative tidal heights, air temperature, and insolation during the May 1997
sampling. Tidal heights unregistered. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Figure 4.
Relative tidal heights, air temperature, and insolation during the September 1997
sampling. Tidal heights unregistered. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Figure 5.
Distribution of salinity within the Chassahowitzka National Wildlife Refuge,
September 1997 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Figure6.
Total Kjeldahl nitrogen as a function of salinity; all samplings and station
categories. (COAST - coastal stations, TCRKS - tidal creeks, HOMOS Homosassa River, SPGS - secondary spring vents, RIVER - Chassahowitzka
River.) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Figure7.
Inorganic Nitrogen:Phosphorus ratios as a function of salinity; all samplings and
station categories. (COAST - coastal stations, TCRKS - tidal creeks, HOMOS Homosassa River, SPGS - secondary spring vents, RIVER - Chassahowitzka
River.) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Figure 8.
Spatial distribution (by latitude) of trophic state indices; all samplings, for
Chassahowitzka riverine (RIVER) and coastal stations (COAST) only. . . . . . . . . 24
Figure 9.
Lack of dependence of trophic state on salinity; coastal stations, all
samplings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Figure10.
Gradient of nitrate-nitrite-nitrogen with respect to salinity; all Chassahowitzka
River samplings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Figure 11.
Gradients of Chlorophyll a, corrected for pheophytin, with respect to
salinity; all Chassahowitzka River samplings. . . . . . . . . . . . . . . . . . . . . . 26
iii
(List of Figures, Continued)
Page No.
Figure 12.
Gradients of Trophic State Index with respect to salinity; all Chassahowitzka River
samplings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Figure 13.
Reverse gradients of organic nitrogen with respect to salinity; all Chassahowitzka
samplings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Figure 14.
Comparison of total nitrogen as a function of salinity, in the Chassahowitzka and
Homosassa Rivers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Figure 15.
Comparison of chlorophyll a, corrected for pheophytin, as a function of salinity,
in the Chassahowitzka and Homosassa Rivers. . . . . . . . . . . . . . . . . . . . . . . . . 28
Figure 16.
Dinoflagellate abundance as function of total phytoplankton concentration; riverine
and coastal stations. Scales truncated. A - May 1996, B - May 1997, C September 1997. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Figure 17.
Riverine gradients of biomass, as the average of “light,” “typical,” and “heavy”
growth, during May 1997. Allocated into vascular submerged aquatic vegetation
(SAV) and macroalgal components. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Figure 18.
Riverine gradients of biomass, as the average of five evenly spaced (across
channel) or randomly placed (open water) quadrats during September 1997.
Allocated into vascular submerged aquatic vegetation (SAV) and macroalgal
components. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Figure 19.
Riverine gradients of sedimentary total Kjeldahl nitrogen and total phosphorus
during May 1997. Rocky substrate and no sample at Stations R4 and 44. . . . . . . 31
Figure 20.
Riverine gradients of sedimentary N:P ratios (N from TKN only) and percent
organics during May 1997. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Figure 21.
Riverine gradient of total Kjeldahl nitrogen and total phosphorus, normalized for
sediment organic content, during May 1997. . . . . . . . . . . . . . . . . . . . . . . . . . 32
Figure 22.
Sediment total Kjeldahl nitrogen content as a function of percent organic content
during May 1997. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
iv
LIST OF TABLES
Page No.
Table 1.
Table 2.
Specific conductance, nitrate-nitrogen concentrations and occurrence of tidal
variation in individual springs. Water quality data are averages of a number of
samplings between 1993 and 1997. (From Jones et al, 1997.) . . . . . . . . . . . . . . . 3
Scalars used for Braun-Blanquet cover-abundance ratings of submerged aquatic
vegetation in Chassahowitzka Bay. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Table 3.
7
Detection limits, analytical methods, and data quality objectives for water quality
samples; Chassahowitzka River and the coastal region. . . . . . . . . . . . . . . . .
8
Table 4.
Geographic locations of all stations, decimal degrees. . . . . . . . . . . . . . . . . .
9
Table 5.
Species list for attached macroalgae and vascular SAV in the coastal area of the
Chassahowitzka National Wildlife Refuge. . . . . . . . . . . . . . . . . . . . . . . . . .
16
Table 6.
Table 7.
Table 8.
Braun-Blanquet estimates of frequency, abundance, and density of the following
species within the coastal vegetated regions in the Chassahowitzka National
Wildlife Refuge. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
18
Average surface or mid-depth observations of physical parameters at 30 coastal
stations, including both morning and afternoon samplings. . . . . . . . . . . . . . .
19
Summary regional nutrient and chlorophyll values for May and September 1997.
Values less than the limit of detection averaged as a zero value. Units are mg/l
for nutrients, ug/l for chlorophyll, mg:mg for nutrient ratios, and unitless for TSI
values. “Riverine” comprised of Chassahowitzka River stations only. . . . . . . .
V
23
ABSTRACT
Shallow water environments offshore of the Chassahowitzka River, west-central Florida, were
sampled for submerged aquatic vegetation and water quality to establish baseline estuarine
conditions in a relatively unstudied national wildlife refuge, and identify indicators of trophic
status. Samplings were a continuation of a program begun in May 1996 and were conducted
during Spring (May 1997) and early Fall (September 1997) to characterize conditions
accompanying historical minimum and maximum periods of freshwater inflow. Stations were also
selected to characterize conditions within the Chassahowitzka and Homosassa Rivers, within tidal
creeks between the two rivers, and at several of the major spring vents of the Chassahowitzka
Spring complex.
Chassahowitzka Bay is a large, shallow (< 2 m), exposed mosaic of saltmarsh, seagrass beds, and
algal assemblages growing on an extensive karstified limestone shelf. Algal and/or seagrass cover
was nearly continuous with moderate densities. Drift algae, followed by Caulerpa paspaloides
were the most common species during the spring samplings with reduced drift algal densities
during the fall sampling of 1997. Other changes in species abundances and density were also
noted.
The water column during the study was well mixed and nutrients, chlorophyll, and trophic state
indices were low, indicating “Good” water quality overall. The northern region apparently
received a higher nutrient loading than the southern region, as indicated by higher trophic state
indices. Algal growth was nitrogen limited in the coastal region and nitrate contributed by the
Chassahowitzka Springs was rapidly removed from the water column within the Chassahowitzka
River. From the main spring to offshore, organic nitrogen in the water column increased with
increasing salinity. Biomass along the riverine gradient was bimodal with minimum values near
the mouth of the Chassahowitzka River in a zone of high salinity variation. Sediment nutrient
concentrations also displayed spatial patterns. The nitrogen content of sediment organics was
relatively consistent, while phosphorus appeared to be preferentially released.
As nitrogen loadings from both groundwater discharge and atmospheric deposition appear to be
increasing, continued monitoring should concentrate on obtaining the data necessary to evaluate
trends in water quality and submerged aquatic vegetation. For trend detection, more frequent
sampling is desirable.
INTRODUCTION
The Chassahowitzka National Wildlife Refuge, under the management of the U.S. Fish and
Wildlife Service, is a 115 square kilometer area that encompasses the mouth and a portion of the
Chassahowitzka River as well as numerous islands and near-shore marine waters out to a distance
of between 1 and 3 kilometers offshore. Located on the central west coast of Florida, this shallow
portion of the Gulf of Mexico is receiving more attention as trends of increasing nutrients in
1
groundwater discharges are documented (Jones et al, 1997; Dixon, 1997). Atmospheric loadings
of inorganic nitrogen in wet deposition have also increased in recent years (Dixon, 1997) at
national monitoring stations in Florida (National Atmospheric Deposition Program; Verna
Wellfield and Bradford Forest). In addition to documenting baseline conditions in the Refuge, the
USFWS has begun a data collection program designed to document trends in water quality and
vegetation which may be attributed to increasing nutrient loads to the region.
Chassahowitzka Bay and adjoining inshore waters are bounded by an extensive system of tidal
wetlands, intertidal shoals and oyster reefs, and include extensive, shallow, subtidal beds of
submerged aquatic vegetation. Geologically, the area represents a mixture of marsh archipelago
and a limestone shelf embayment situated along a low energy, low-gradient, micro-tidal, and nonbarrier coast (Hine and Belknap, 1986). Coastal hydrology is dominated by the discharge of
several spring groups (Homosassa, Hidden River, Halls River, Chassahowitzka). The
Chassahowitzka complex accounts for approximately 7.58 cubic meters per second, while the
various springs contributing to the Homosassa River total nearly 10.02 cubic meters per second
(Jones et al., 1997). Discharges are greatest in September and October, and least in June and July
(Yobbi, 1992). Spring flows from a groundwater basin of approximately 492 square kilometers
are supplemented by runoff from a poorly delimited coastal basin area of about 800 square
kilometers. Rainfall (142 cm per year average) is greatest from May through September and least
in November (Yobbi, 1992). Depending on their depths, individual spring vents’ discharge
variable mixtures of fresh and saltwater and both the quality and quantity of discharge can vary
tidally (Yobbi, 1992). Summarized from Jones et al. (1997), the specific conductance, nitratenitrogen concentrations and whether water quality varies tidally appear in Table 1 for individual
named springs of the Chassahowitzka and Homosassa systems. Water levels at the
Chassahowitzka main spring also vary tidally and are reported in real time by the Tampa District
Office of the U.S. Geological Survey (http://www-tampa.er.usgs.gov/Realtime/data/023 10650_
gh.html).
2
Table 1.
Specific conductance, nitrate-nitrogen concentrations and occurrence of tidal
variation in individual springs. Water quality data are averages of a number of
samplings between 1993 and 1997. (From Jones et al, 1997.)
Spring
Chassahowitzka
Chassahowitzka #1
Chassahowitzka Main
Crab Creek
Baird
Ruth
Potter Creek
Beteejay Head Spring
Blue Run
Homosassa
Abdoney
Belcher
Trotter #l
Trotter Main
McClain
Pumphouse
Homosassa River #l
Homosassa Main #l
Homosassa Main #2
Homosassa Main #3
Hall’s River Head
Hidden River Head
Hidden River #2
Specific
Conductance
(umhos/cm)
N03-N
(mg/l)
Tidal Variation
In Quality
(Yes/No)
770
1730
5294
11426
2977
9770
671
7747
0.47
0.42
0.46
0.21
0.44
0.33
0.27
0.26
N
Y
Y
Y
Y
Y
N
Y
489
0.46
0.44
0.44
0.42
0.36
0.39
0.39
0.43
0.42
0.44
0.32
0.66
0.53
N
N
N
N
N
N
Y(?)
Y
Y
Y
Y
Y(?)
Y(?)
455
462
429
548
506
4160
3245
5694
1339
7297
1449
2162
Major ion chemistry of Homosassa and Chassahowitzka springs is dominated by sodium chloride,
with more calcium and bicarbonate in the Chassahowitzka (Estevez et al., l991 ; Jones et al.,
1997). A groundwater and spring discharge monitoring program together with other trend
analyses conducted by the Southwest Florida Water Management District (SWFWMD, 1994;
Jones et al., 1997; Dixon, 1997) has documented increasing trends of nitrogen in spring
discharges with sources attributed to inland development and subsequent residential and golf
course fertilization (Jones et al., 1997). Nitrate levels in the discharge are presently near 0.4 0.5 mg/l or between 40 and 50 times background groundwater concentrations. Concentrations of
total phosphorus are typically near analytical detection limits (Yobbi, 1992).
3
Coastal plant communities are dominated by hammock forest, the wettest of which (with live oak
[Quercus virginiana], southern red cedar [Juniperus silicicola] , and cabbage palm [Sabal
palmetto]) intergrades with marshes of sawgrass (Cladium jamaicense) and saltmarsh dominated
by black needlerush (Juncus roemerianus). Cattail (Typha spp.) and giant reed (Phragmites
australis) also occur in the fresher regions. Spring runs punctuate the hammock forest and are
fringed by mixed hardwoods, bald cypress (Taxodium distichum) (Wolfe, 1990), and numerous
sedge, rush, and grass marshes intolerant of saltwater. Clear, mineralized flows in the spring runs
permit luxuriant growth of tape grass (Vallisneria neotropicalis). Other common submerged fresh
to brackish water species include sago pondweed (Potamogeton pectinatus), watermilfoil
(Myriophyllum spicatum), hydrilla (Hydrilla verticillata), and naiad (Naja guadalupensis) , with
Potamogeton illinoensis occurring near the spring head. Floating rafts of both Potamogeton and
Myriophyllum mixed with Enteromorpha spp. and filamentous algae occur at times in the upper
river, particularly in areas of low current velocity. Filamentous forms of green and blue-green
algae are also common with the blue green alga, Lyngbya spp., at nuisance levels in the main
spring.
The near shore bottom is shallow and low in relief. Bottoms are limestone covered by veneers
of organic and carbonate sediments lacking quartz sand (Hine and Belknap, 1986). In the coastal
area, submerged aquatic vegetation (SAV) is extensive. Contiguous beds of dense SAV cover
more than 90 % of inshore ( < 2 m depth) areas (McNulty et al., 1972; Wolfe, 1990). Dominant
vascular plants are turtle grass (Thalassia testudinum), manatee grass (Syringodium filiforme), and
shoalgrass (Halodule wrightii) (Iverson and Bittaker, 1986). Widgeon grass (Ruppia maritima)
grows in lower salinity inshore waters. Planktonic and benthic algal communities of the
immediate area are unstudied but reports from nearby, similar areas depict a rich, abundant flora
(Phillips, 1960; Earle, 1969; Dawes, 1974). The bulk of the waters within the study area are
designated as Class II (Shellfish propagation or harvesting, F.A.C. 17-302) by the Florida
Department of Environmental Protection (FDEP).
The present study was performed to describe SAV and water chemistry during a dry and a wet
season; to establish a 1996-97 baseline of ecological conditions; and to report on existing
indicators of eutrophication in the near coastal waters along an undeveloped area, the
Chassahowitzka National Wildlife Refuge (Figure 1).
MATERIALS AND METHODS
In 1997, sampling dates were chosen in May and September in order to sample historically dry
and wet conditions. The May sampling also provided continuity with the May 1996 sampling.
Sampling for SAV took place from May 19 through May 21, 1997, with methods described below.
Water quality sampling took place during the early morning of May 21 and was repeated during
the late afternoon of the same day. The wet season sampling took place September 15-17, 1997
with the water quality run conducted the early morning and late afternoon of September 17.
Sampling times were selected based on observed patterns of hypoxia and supersaturation in other
4
water bodies (Marshall and Leverone, 1994). Mote Marine Laboratory (MML) conducted all
sampling and analysis with sampling assistance from Refuge personnel and vessels.
For coastal SAV, the study area encompassed a 115 square kilometer coastal area. Most stations
were in open waters between Homosassa Point (28°46.5'N) to the north, to Raccoon Point
(28 °36.3'N), and seaward of saltmarshes and the extensive tidal creek system. Stations extended
out to the entrance marker for Chassahowitzka (82°44.3'W).
Sampling for SAV in 1997 revisited all 30 stations established in May 1996 (Figure 1). During
1996, station selection was performed with assistance from the U.S. Environmental Protection
Agency (Dr. Kevin Summers) following EMAP (Environmental Monitoring and Assessment
Program) protocols which allow the unbiased estimations of the area1 extent of a selected level of
a given parameter (i.e., the percentage of a region with dissolved oxygen values greater than
5.O mg L-l). The Refuge was gridded into fifty 1 km2 hexagons and a series of random stations
identified within each polygon. For SAV, the study area was limited to the near coastal waters
outside of the mouths of the Chassahowitzka and Homosassa Rivers. Polygons were eliminated
from consideration if they consisted of predominantly land rather than open water, as a study of
vegetation in tidal creeks was not intended. From the area bounded by the Refuge, 30 polygons
were selected for SAV sampling. Of the potential stations identified within the polygon, sites were
visited in order until SAV was observed.
Cover and abundance of submerged aquatic vegetation (SAV) were measured using a rapid-survey
technique (Braun-Blanquet, 1932). At each station, an observer on each side of the vessel
surveyed two quadrats, each 0.25 m2 in area. Each observer listed species and major plant groups
(seagrass; drift algae, etc.) and assigned a cover-abundance value for each taxon or group
(Table 2). Upper scale values (5-2, inclusive) pertained to cover only. Other scale values were
estimators of abundance (number of individuals per species). Four replicate quadrats were used
to assess within and between station variability. Attributes were calculated as follows:
Frequency =
Abundance =
=
Density
No/Nt
sum of B-B scale values/No
sum of B-B scale values/Nt
where No is the number of occupied quadrats and Nt is the total number of quadrats. The drift
algae were lumped into a single category of miscellaneous algae with frequency, abundance, and
density estimated as for the attached species. Secchi depths were recorded during SAV surveys
and underwater photographs secured at most stations during both May and September. During
the September sampling, canopy height was also estimated by species to differentiate between
varying morphologies, growth habits, and resultant biomass of a given species.
5
Figure 1.
Chassahowitzka National Wildlife Refuge, Chassahowitzka, Florida. Thirty
coastal stations, secondary spring vents sampled (CH#l, CRAB, POTTR),
meteorological station (MET), and water level station (WL).
6
On September 16, 1997, a researcher (Mr. Joe Conti) performing similar B-B surveys in the St.
Martin’s Marsh area to the north of the Refuge and the Homosassa River accompanied MML
crews and cross-calibrated techniques and species identifications. During May and September,
stations in the Chassahowitzka River and coastal Stations 24, 26, and 44 were sampled for wet
2
weight biomass and percentage of attached vascular SAV. During May, quadrats (0.25m ) for
biomass were placed in areas of light, typical, and heavy growth, while during September, five
Braun-Blanquet quadrats were sampled at either randomly (in the coastal region) or at equal
distances across the width of the river.
Table 2.
Value
5
4
3
2
1
+
r
Scalars used for Braun-Blanquet cover-abundance ratings of submerged
aquatic vegetation in Chassahowitzka Bay.
Relative Number
Percent Cover
Remarks
any number
any number
any number
any number
numerous
few
solitary
greater than 75%
50%-75%
25%-50%
5%-25%
less than 5%
small
small
- or scattered with up to 5%
- assigned 0.5 value
- assigned 0.1 value
During May and September, in situ parameters were collected in the early morning and late
afternoon at all 30 SAV stations. A subset of 20 polygons (generally those nearer shore) was
sampled during the morning for water quality (nutrients, chlorophyll, and phytoplankton). In situ
samplings consisted of measurements of depth, temperature, conductivity (salinity), dissolved
oxygen, and percent saturation of oxygen. Instrumental measurements were made at near surface
and near bottom unless depths were less than 1 m, in which case only a middepth reading was
made. Water quality analyses consisted of nutrients (nitrate-nitrite-nitrogen, ammonium-nitrogen,
total Kjeldahl nitrogen, orthophosphate, total phosphorus), chlorophylls (chlorophyll a, b, c,
pheophytin, and chlorophyll a corrected for pheophytin), and phytoplankton enumeration.
Analytical methods and detection limits were the same as used for the 1996 sampling and are
detailed in Table 3. Instrumental readings, collection, preservation and analysis of water quality
samples were performed according to MML’s Florida Department of Environmental Protectionapproved Comprehensive Quality Assurance Plan. Trophic state indices were calculated according
to the methods of Hand et al. (1988), as the average of individual percentiles for a limiting
nutrient, chlorophyll, and Secchi depth, with a “GOOD” rating equivalent to a value of 50 or lower
and analogous to a chlorophyll concentration of 10 ug/l. Indices were computed without Secchi
percentiles when Secchi depths were greater than bottom, and numeric values of one half of the
limit of detection used when sample concentrations, were less than the limit of detection. Samples
7
for phytoplankton enumeration were preserved with Utermohl’s solution (potassium iodide, iodine,
and acetic acid). Both total organisms and total dinoflagellates were counted using twelve 1.00 ml
aliquots in Sedgewick-Rafter counting chambers under 100X magnification. Dinoflagellates,
diatoms, and cysts were counted separately as increasing nitrogen nutrient concentrations favor
the growth of flagellates (Officer and Ryther, 1980).
Table 3.
Detection limits, analytical methods, and data quality objectives for water quality
samples; Chassahowitzka River and the coastal region.
Parameter
Detection
Limit
Method
Number
Ortho-phosphorus
Total Phosphorus
Ammonium Nitrogen
Nitrate-Nitrite-Nitrogen
Total Kjeldahl Nitrogen
Chlorophylls
0.005 mg/l
0.05 mg/l
0.005 mg/l
0.005 mg/l
0.05 mg/l
0.5 ug/l
4500-PF
365.4
350.1
353.2
351.2
10200H (1,2)
Maximum
Precision
(%RSD)
Recovery
Limits
(%)
15
13
20
15
18
28
87-l 15
80-111
86-l 13
81-116
86-119
Not Appl.
Seven salinity-based stations in the nine kilometer run of the Chassahowitzka River (including the
main boil), were also sampled for in situ and water quality parameters, as were six stations in the
13 km run of the Homosassa River, and five in several of the tidal creeks between the two rivers.
Additional water quality samples were collected from three of the smaller individual spring vents
contributing to the Chassahowitzka system. Locations of all stations appear in Table 4.
Additional parameters measured at the springs during May, 1997, included calcium, sodium,
magnesium, and iron. The Homosassa, tidal creek, and spring stations were sampled for both
physical and chemical parameters, but were not sampled for SAV. Samples collected within the
riverine portions of the Chassahowitzka and the Homosassa Rivers were combined with selected
stations offshore to examine gradients in water quality with respect to salinity (Figure 2).
Incident photosynthetically active radiation (PAR) and air temperature were recorded continuously
for the duration of the sampling at the maintenance facility of the Chassahowitzka National
Wildlife Refuge (Figure 1, Station MET) using a LiCor 1000 (with 190SB) and a Ryan
Tempmentor, respectively. Temperature readings were made approximately 0.6 m above grade.
Relative tidal heights were recorded during the sampling periods, but do not permit inter-sampling
comparisons as instrument installations were not registered to a vertical datum. Tide levels were
recorded with an ISCO 4230 bubbler flow meter, located at 28°42.39'N, 82°37.12'W on the
eastern bank of Little Gator Creek, just off the main stem of the Chassahowitzka River (Figure
1, Station WL).
8
Geographic locations of all stations, decimal degrees.
Table 4.
Station
Latitude
Decimal Deg.
Longitude
Decimal Deg.
Station
Latitude
Decimal Deg.
Longitude
Decimal Deg.
Chassahowitzka Coastal Area
Tidal Creeks
PETTY
28.7660700
-82.6448334
2
28.7697619
-82.7136666
7CABBG
28.7312645
-82.6567223
3
28.7674873
-82.7004434
ROSE
28.7361533
-82.6446666
4
28.7603473
-82.6927508
MASON
28.7358200
-82.6449167
6
28.7552940
-82.6702202
UMASON
28.7603756
-82.6263056
7
28.7474566
-82.7004702
8A
28.7494843
-82.6612509
Homosassa River
HO
28.7688478
-82.6448334
9
28.7417113
-82.7195300
HO.5
28.8005280
-82.5958890
12
28.7281232
-82.7094717
H1
28.7955978
-82.6049166
14
28.7202912
-82.7026642
H2
28.7854866
-82.6190000
17
28.7116276
-82.7088333
H3
28.7835422
-82.6278889
18
28.7121534
-82.6982493
H4
28.7751256
-82.6758056
19
28.7078457
-82.7127493
20
28.7024598
-82.6605267
Other Spring Vents
POTTR
28.7316670
-82.5972780
21
28.6985438
-82.6778323
CRAB
28.7174440
-82.5752780
22
28.6969613
-82.6999176
CH#l
28.7155560
-82.5749170
23
28.6930399
24
28.6841511
-82.6546688
Chassahowitzka River
-82.6676668
RO
28.7157922
-82.5759167
25
28.6809056
-82.6788623
R1
28.7148478
-82.5765833
26
28.6748492
-82.6654727
R1.3
28.7161229
-82.5870556
27
28.6611806
-82.6733638
R1.7
28.7204843
-82.5992758
28
28.6239569
-82.6729453
R2
28.7142938
-82.6086689
30
28.6152076
-82.6583058
R2.5
28.7159866
-82.5871389
39
28.7460726
-82.6659716
R3
28.7026256
-82.6236666
4OC
28.6445134
-82.6735569
R4
28.6934589
-82.6405000
41
28.7613183
-82.6882500
42
28.7537383
-82.6868606
44
28.6605959
-82.6865281
45
28.6488478
-82.6878316
46
28.6305713
-82.6751662
47
28.6177342
-82.6778913
Figure 2.
Chassahowitzka and Homosassa River gradient stations and the tidal creek stations
at Seven Cabbage Cutoff (7CABG), Rose Creek (ROSE), Mason Creek
(MASON), upper Mason Creek (UMAS), and Petty Creek (PETTY).
10
During both May and September samplings, sediment samples along the Chassahowitzka riverine
transect were collected as a single surface grab (generally the top 2-5 cm) and were processed for
grain size distribution (as % volume of various size classes) with a Coulter LS200 laser particle
sizer. Analysis of sediments included the fraction greater than 2 mm in diameter. During May,
riverine sediments were collected and processed for total phosphorus, total Kjeldahl nitrogen, and
percent organics by combustion.
On November 5, 1997, a survey of Thalassia was conducted on plants which appear to consist of
a single rhizome growing linearly in a fracture feature of a shallow limestone platform. Visited
at low tide, Thalassia samples were excavated for examination of shoot age and morphology. Dye
was used to determine that there was no obvious groundwater discharge from the fracture during
the collection effort. Fracture orientation and length were mapped for comparison with other
geological information of the area.
In other opportunistic surveys of the Chassahowitzka River and Refuge, a partial listing of
molluscs was identified from living specimens and shell material retained from qualitative
sediment samples by a 2mm sieve. The sampling was conducted on February 17, 1998 at nine
riverine and coastal stations and was accompanied by collection of in situ water quality
parameters.
RESULTS AND DISCUSSION
Study Area
The largest discharge for the Chassahowitzka River appears to be the main boil (Station RO)
immediately to the northwest of the Citrus County boat launching facility. Numerous smaller
vents were observed immediately upstream (including Chassahowitzka #l- Station CH#l) and
some very small discharges were noted during the 1996 sampling within some of the manmade
canals adjacent to the spring head. At that time, the upstream discharges were lower in
conductivity than the main springs, indicating source water from a more surficial aquifer.
Immediately downstream of the main spring area, the confluence with Crab Creek delivered the
discharge from another sizeable vent (Station CRAB), with a conductivity higher than that of the
main spring. Station R1 was immediately below this confluence, representing both discharges.
Floating mats of senescent filamentous vegetation and a few cattails (Typha sp.) lined the sides of
the river at this point, but the main stream was clear with a cloudy-blue appearance characteristic
of many groundwater discharges. Beds of tapegrass (Vallisneria) and pondweed (Potamogeton
pectinatus, Potamogeton illinoensis) were visible on the bottom with some Hydrilla, as well. Tidal
changes in water elevation, but no reversing flows, were observed at this site.
Baird Creek and Potter Creek also contribute spring flows to the River. Stations R1.3 and Station
R1.7 are located immediately downstream of each confluence. (Discharge from the spring at the
head of Potter Creek was sampled as Station POTTR.) The upper 4 kilometers of the river are
surrounded by a deciduous flood plain forest down to the eastern boundary of the Chassahowitzka
11
National Wildlife Refuge (Station R2). At the boundary, sawgrass (Cladium jamaicensis) and
cattail dominate the bank vegetation, cabbage palm(Sabal palmetto) hammocks and Juncus appear,
and the flood plain forest ends. Floating mats of enteromorpha-like algae, Eurasian water milfoil
(Myriophyllum spicatum), and Hydrilla verticillata are very dense here. Water clarity typically
decreases by this location and has acquired the brown color of dissolved humics. Reversing tidal
flows are observed at this station.
By just upstream of Crawford Creek (Station R2.5) and at Dog Island (Station R3), sawgrass lines
the river, while Juncus dominates the marsh at higher elevations. Some cattails remain and the
seagrass Ruppia maritima is present in the river. Water clarity at Station R3, as at R2, was
generally fairly low, presumably as humics precipitated with increasing salinity. At Station R4,
opposite Johns Island, the marsh was dominated by Juncus, some Spartina alterniflora was
present, and sawgrass ended. Water clarity was much improved by this station, and the remaining
coastal waters sampled were noticeably clear. The remaining coastal stations were within and
offshore of a dense archipelago of marsh (Juncus) islands. Oyster bars were observed in some
tidal creeks but were not a dominant land form. Mangroves (Rhizophora mangle) were not
numerous, being near the northern limits of cold tolerance for this species, and were evident
chiefly as either dead and weathered stumps, or as small seedlings in the 20-50 cm size range.
Offshore stations used to complete the riverine gradient analyses consisted of the coastal stations
24, 26, and 44.
Tidal creek stations located between the Chassahowitzka and Homosassa Rivers, were located in
Seven Cabbage Cutoff (7CABG), Petty Creek (PETTY), Rose Creek (ROSE), and near the mouth
and upstream in Mason Creek (MASON, UMASON). Development on the land surrounding each
of these stations appears minimal with the exception of Mason Creek, where a medium density
residential land use appears on the north bank of upper Mason Creek. Juncus marshes, followed
by low elevation karst islands with cedars and palms are the dominant vegetation and land form
apparent adjacent to the tidal creek stations.
Stations sampled in the Homosassa River extended from the limit of boat traffic near Homosassa
Main (Station HO), to the mouth of the river. Forested wetlands lines both sides of the river down
to approximately Station H2. Residential development and recreational water-based industries are
mush more extensive than along the Chassahowitzka. Portions of the upper river are sea-walled.
Downstream of Station H2, Juncus marshes become more prevalent, with frequent limestone
platforms of low relief supporting cedars and cabbage palms. The limestone platforms are
analogous to Johns Island on the Chassahowitzka but are much more numerous. Residential
development becomes less frequent but is still a noticeable presence.
Weather
May 1997 was extremely dry, with the Northern Coastal Basin of the Southwest Florida Water
Management District reporting a rainfall of 1.8 cm; or a deficit of 7.6 cm below the historical May
mean of 9.4 cm. At this time, annual deficits from 12-month historical means totaled 30.9 cm
(SWFWMD, 1997a). Of the 16 Florida aquifer wells in the northern region monitored by
12
SWFWMD, all water level elevations were lower than in the prior May (SWFWMD, 1996) as
well as being below historical means for May by between 25 cm and 2.2 m. In September, rainfall
totaled 12.8 cm, a 3.3 cm deficit from the historical September value, and the annual deficit had
reached 42.2 cm (SWFWMD, 1997b). Groundwater elevations were between 17 cm and 1.9 m
lower than in September 1996, and were 37 cm to 2.8 m lower than historical averages for
September.
The solar irradiance and air temperature a short distance inland (Station MET, Figure 1), and tide
heights within the river (Station WL, Figure 1) as recorded during the sampling days appear in
Figures 3 and 4. During May, the weather was fair with scattered clouds, clearing for the
afternoon water quality run on the final day. Inland air temperatures were moderate (18-40°C).
During September, the initial sampling days were very clear with increasing cloudiness. Air
temperatures ranged from 20-45 ° C . During the last day, a severe thunder storm late in the
afternoon forced abandonment of the last water quality stations (Station 8A, 39). Water quality
samplings in May 1997 took place on a falling and low tide in the early morning, followed by the
afternoon sampling centered around a high tide. Tidal range was less than 0.75 m (2.5 ft) with
a predicted high of 0.94 m. Water quality data were collected on very comparable tidal phases
during September 1997, but with predicted highs of 0.83 m. Observed tidal range was also
somewhat smaller during September than in May, approximately 0.61 m (2 ft). Peak insolation
was also lower during September.
SAV Sampling
Of the 30 polygons identified for sampling in 1996, vegetation was found at all but one of the
randomly chosen primary sites within each polygon. For Polygon 8, the primary site had no
vegetation and the first alternate site, Station 8A, was sampled instead. Data on frequency,
abundance, and density were computed on the 30 vegetated stations, and so represented an
assessment of the vegetated regions only, rather than an assessment of the entire submerged study
area. Assessments of the entire study area would employ data from the four unvegetated quadrats
of Station 8 rather than the data from Station 8A, but were not presented. The fact that only one
of the 30 primary stations was unvegetated indicates that coverage by SAV is extensive and almost
continuous. (Station 40C was the primary site for Polygon 40, as the other randomly selected sites
fell on land.) Sampling during 1997 revisited the 30 vegetated stations sampled during 1996, and
so again represents an assessment of the vegetated regions.
Sampling of SAV was carried out at the 30 vegetated stations during May and September 1997 and
individual quadrat data appear in Appendix A. Reported depths were observed and are not tidally
corrected. Secchi depths during May 1997 were all greater than the water depth. During
September, four stations in the northern portion of the study area (Station 3, 41, 6, and 39)
recorded Secchi depths at or less than total water depths. Species encountered during 1996 and
1997 appear in Table 5 and include representatives of the algal phyla Chlorophyta and Phaeophyta,
as well as marine angiosperms or seagrasses.
13
Figure 4. Relative tidal heights, air temperature, and insolation during the September 1997
sampling. Tidal heights unregistered.
15
Table 5.
Species list for attached macroalgae and vascular SAV in the coastal area of the
Chassahowitzka National Wildlife Refuge.
May-96
May-97
Sep-97
Acetabularia crenulata
Acetabularia crenulata
Acetabularia crenulata
Batophora oerstedi
Batophora oerstedi
Anadyomene stellata
Caulerpa ashmeadii
Caulerpa paspaloides
Avrainvillea spp.
Caulerpa paspaloides
Caulerpa prolifera
Batophora oerstedi
Caulerpa prolifera
Chara spp.
Caulerpa paspaloides
Halimeda incrassata
Halimeda incrassata
Caulerpa prolifera
Halodule wrightii
Halodule wrightii
Chara spp.
Penicillus capitatus
Halophila englemannii
Digenia simplex
Ruppia maritima
Penicillus capitatus
Halimeda incrassata
Sargassum spp.
Ruppia maritima
Halodule wrightii
Thalassia testudinum
Syringodium filiforme
Halophila englemannii
Udotea flabellum
Thalassia testudinum
Penicillus capitatus
Udotea flabellum
Sargassum spp.
l
Thalassia testudinum
Udotea flabellum
Anadyomene stellata, Avrainvillea spp., Chara spp., Digenia simplex, Halophila englemannii, and
Syringodium filiforme were recorded in the quadrats for the first time in 1997, while Caulerpa
ashmeadii was not found in 1997. Those attached species occurring most frequently were as
follows :
May 1996
May 1997
Sept 1997
Caulerpa paspaloides, Acetabularia, Batophora, Halodule, and Thalassia
Caulerpa paspaloides, Halodule, Batophora, Acetabularia, and Thalassia
Caulerpa paspaloides, Halodule, Thalassia, Penicillus, and Batophora
The most frequently observed attached species in all three samplings to date (Table 6) was
Caulerpa paspaloides, recorded in 38 %, 36 %, and 37 % of all quadrats in May 1996, May 1997,
and September 1997, respectively. Acetabularia was observed approximately half as often in 1997
compared to May 1996. The September sampling recorded more instances of Penicillus,
Halodule, and Thalassia than during either of the previous May events. Comparable numbers of
species were observed during each sampling. Although not quantified, no problems of excessive
colonization by epiphytes were noted during the surveys.
16
Where it occurred, Caulerpa paspaloides averaged a B-B abundance value of 3.2 to 3.3 in 1997,
or 25-50 % coverage. This abundance was a decrease from that observed in 1996, when coverage
was estimated at 50-75 % . In general, the five most frequent species declined in abundance, both
between May 1996 and 1997, and between May and September 1997. Within the vegetated study
area, Caulerpa paspaloides has consistently been the attached species with the densest coverage,
near 5% coverage overall. Data by station and major species of Braun-Blanquet estimates of
frequency, abundance, and density also appear in Appendix A.
In addition to attached algal and seagrass species, drift algal species were also quite common.
During May 1996 and 1997, drift material was observed in 49 % and 43 % of all quadrats,
respectively. By September, however, drift species were only observed in 13 % of quadrats.
During May 1997, drift species were similar to those observed during May 1996; Chondria
tenuissima, Laurencia poitei, and Spyridia filamentosa. In addition, Digenia simplex and
Anadyomene stellata were collected as unattached species during the May 1997 sampling. In
September 1997, drift species consisted primarily of Laurencia poitei and Digenia simplex. The
attached but chlorotic Caulerpa paspaloides observed during 1996 was not present during the 1997
samplings.
Water Chemistry - Physical Parameters
Summary values of in situ parameters for all three samplings to date appear in Table 7 with
individual data from 1997 contained in Appendix B. The shallow coastal waters near the mouth
of the Chassahowitzka River were well mixed during both May and September samplings.
Average surface to bottom differences in salinity were less than 0.1 PSU with maximum
differences less than 1.6 PSU for all three samplings. Overall, September is noted for higher
salinity values and lower concentrations and percent saturation of dissolved oxygen.
17
Table 6.
Braun-Blanquet estimates of frequency, abundance, and density of the following species within the coastal vegetated regions
in the Chassahowitzka National Wildlife Refuge.
Frequency (%)
Species
Abundance (BB Units)
Density (BB Units)
May-96 May-97 Sep-97 May-96 May-97 Sep-97 May-96 May-97 Sep-97
Caulerpa paspaloides
38
36
37
4.0
3.3
3.2
1.5
1.2
1.2
Acetabularia crenulata
31
17
12
2.1
2.2
0.6
0.7
0.4
0.1
Batophora oerstedi
18
18
13
0.8
1.3
0.6
0.2
0.2
0.1
Halodule wrightii
18
20
26
3.8
2.9
2.6
0.7
0.6
0.7
Thalassia testudinum
18
15
23
3.2
2.9
2.9
0.6
0.4
0.7
Caulerpa prolifera
13
11
11
2.6
2.2
1.6
0.3
0.2
0.2
Penicillus capitatus
8
8
21
1.9
0.6
1.5
0.2
0.0
0.3
Ruppia maritima
5
5
0
3.1
1.7
0.0
0.2
0.1
0.0
Sargassum spp.
5
0
7
3.5
0.0
1.5
0.2
0.0
0.1
Udotea flabellum
3
3
3
0.2
0.3
0.2
0.0
0.0
0.0
Halimeda incrassata
3
12
6
0.7
1.4
0.5
0.0
0.2
0.0
Caulerpa ashmeadii
1
0
0
2.0
0.0
0.0
0.0
0.0
0.0
Anadyomene stellata
0
0
3
0.0
0.0
0.1
0.0
0.0
0.0
Avrainvillea spp.
0
0
3
0.0
0.0
0.9
0.0
0.0
0.0
Chara spp.
0
7
1
0.0
3.5
3.0
0.0
0.2
0.0
Digenia simplex
0
0
8
0.0
0.0
2.4
0.0
0.0
0.2
Halophila englemannii
0
12
3
0.0
3.4
1.3
0.0
0.4
0.0
Syringodium filiforme
0
3
0
0.0
3.3
0.0
0.0
0.1
0.0
Misc. Algae
49
43
13
3.6.
3.5
2.4
1.8
1.5
0.3
Unvegetated
3
0
3
5.0
0.0
5.0
0.1
0.0
0.1
18
Non-parametric testing (Kruskal-Wallis) identified a number of significant differences between the
three sampling dates, with May 1996 to May 1997 varying significantly only with respect to
temperature, while May to September 1997 recorded significant differences (p < =0.05) in all four
physical parameters.
Table 7. Average surface or mid-depth observations of physical parameters at 30 coastal
stations, including both morning and afternoon samplings.
Salinity
(PSU)
May 1996
May 1996
Sept 1997
20.5
20.7
26.0
Temperature
(Deg C)
Dissolved Oxygen
(mg/l)
27.1
28.4
29.4
6.8
6.6
5.5
Percent Saturation
%
97
96
83
Despite the effort to sample a wet season in September 1997, rainfall and groundwater deficits
were reflected in the increased salinities observed during this month, as compared to the prior two
May samplings. Predicted tidal heights were lower in September with no major weather patterns
noted which would have elevated salinity values. During September, the stations with higher
salinity were located in the southern portion of the Refuge (Figure 5). Salinity in May 1996
ranged between 12 and 27 PSU in the coastal stations, 14 to 28 PSU during May 1997, and 21
to 32 PSU during September 1997.
Violations of Florida water quality criteria occurred at several locations. During the early
morning samplings of May 1996, May 1997, and September 1997, three, two, and 16 stations,
respectively, had surface dissolved oxygen values less than the 4.0 mg/l instantaneous criteria.
During September 1997, the majority of stations recording DO values below 4.0 mg/l were in the
northern portion of the Refuge. A D. O. value of 1.9 mg/l was recorded at Station 20, to the north
of the mouth of the Chassahowitzka River. In addition to low morning D. O. values, the average
of both morning and afternoon values of both D.O. and percent saturation were also lower during
September than during the May samplings. Percent saturation values account for differences in
salinity and temperature between samplings by evaluating oxygen present against what is
theoretically possible under thermodynamic laws. Although some portion of the low D.O. values
during the afternoon may be attributed to lower insolation values, storm conditions, and the
longer sampling during September, a comparison of data collected during the peak saturation
period (1500-1730 hours) still identified September as having lower percent saturation values
(108% versus 125% in May 1997).
19
Figure 5.
Distribution of salinity within the Chassahowitzka National Wildlife Refuge,
September 1997.
20
Changes in the diurnal range of dissolved oxygen and percent saturation were also observed
between the three samplings. In addition to the depressed absolute values of dissolved oxygen in
September, the range in saturation was also less during this period. Early morning to late
afternoon percent saturation varied by 69 % , 60 % , and then 48 % for the coastal stations in May
1996, May 1997, and September 1997, respectively, although the difference between the 1997
samplings was not statistically significant for the coastal stations.. Riverine stations expressed
similar decreases in September, with the range in percent saturation approximately half of that
observed for the two prior samplings. While not conclusive, the sum of the B-B abundance values
at the coastal stations followed the same temporal, distribution, highest during the first May
sampling, and lowest during September. Other efforts to link station dissolved oxygen data (either
as concentration, percent saturation, or diurnal range) to abundances at individual stations were
not successful, due in part to the extremely heterogeneous nature of the coastal vegetation, and
comparatively strong tidal currents, which lessen the residence time of any particular water mass
over a particular site.
For riverine stations, stratification was also minimal, with average surface to bottom differences
in salinity of less than 0.5 PSU for all three samplings. A maximum value of 4 PSU was observed
at Station 1.7 during September, and other instances of stratification (of about 2 PSU) were
occasionally seen at Station 4, near the mouth of the River where the riverbed is deeper. Data
from riverine stations were consistent with the temporal trends seen in the coastal water mass.
Dissolved oxygen and percent saturation were lower in September 1997 than previously observed,
while salinity was increased from 5 to 10 PSU on average within the River. Oxygen data from
the Homosassa River are not strictly comparable with data from the Chassahowitzka, as it was
collected mid to late morning and only once during the day.
The tidal creeks exhibited a range of salinities with PETTY, UMAS, MASON, ROSE, and
7CABG ordered from freshest to most saline. Elevated September salinities were evident at these
locations also. Salinity values ranged from 8 to 17 PSU during May 1997, and from 16 to 25 PSU
during September. The tidal creek stations were sampled during early to mid-morning. Percent
saturation values ranged between 45-75 % with only one dissolved oxygen violation recorded, at
7CABG during May 1997 (3.2 mg/l).
Elevated September salinities were also apparent in the discharges from the other spring vents of
the Chassahowitzka system. The CRAB station ranged from 3.5 to 4.6 PSU, POTTR from 3.8
to 9.8 PSU, CH#l from 0.2 to 1.0 PSU, and the headspring, RO, from 0.6 to 3.2 PSU. The most
upstream Homosassa station (HO), although not located directly over a vent, ranged between 0.6
and 2.5 PSU. In each instance, the highest salinity was recorded during the September sampling.
Diurnal variations in salinity were also observed at each vent sampled directly. Dissolved oxygen
was typically low at most of the vents, consistent with undersaturated groundwater. POTTR, the
most saline of the vents, regularly recorded D.O. values below 4.0 mg/l. While some diurnal
variation in percent saturation was observed, extensive forest canopy at most of these locations
appeared to inhibit wide diurnal swings in oxygen, despite extensive rooted and algal vegetation.
21
Water Chemistry - Nutrients
Water quality is characterized by a strong salinity control and a stable riverine gradient. All
analytical results for the 1997 samplings appear in Appendix C, with summary values for coastal
and riverine stations below in Table 8. Coastal stations were distinguished by comparatively low
levels of inorganic nitrogen species, especially nitrate-nitrite-nitrogen. Inorganic or orthophosphorus levels were also low, but generally detectable ( > 0.005 mg/l), while total phosphorus
concentrations were less than the analytical limit of detection (0.05 mg/l) in 75 % of the 1997
samples. Total nitrogen values were moderate and increased with increasing salinity (Figure 6),
in both spatial and temporal scales and among all station categories: Inorganic N:P ratios indicate
the coastal waters were strongly nitrogen limited (values less than 10) during May 1997, while
increases in ammonia and nitrate-nitrite-nitrogen in September produced a more balanced system
(ratios between 10 and 30), albeit with higher inorganic nutrient concentrations overall. Figure
7 illustrates the inorganic N:P ratios for the various station categories of all three samplings to
date.
Figure 6.
Total Kjeldahl nitrogen as a function of salinity; all samplings and station
categories. (COAST - coastal stations, TCRKS - tidal creeks, HOMOS Homosassa River, SPGS - secondary spring vents, RIVER Chassahowitzka River.)
22
Figure 7.
Table 8.
Inorganic Nitrogen:Phosphorus ratios as a function of salinity; all
samplings and station categories. (COAST - coastal stations, TCRKS tidal creeks, HOMOS - Homosassa River, SPGS - secondary spring vents,
RIVER - Chassahowitzka River.)
Summary regional nutrient and chlorophyll values for May and September 1997.
Values less than the limit of detection averaged as a zero value. Units are mg/l for
nutrients, ug/l for chlorophyll, mg:mg for nutrient ratios, and unitless for TSI values.
“Riverine” comprised of Chassahowitzka River stations only.
Coastal 5/97
Coastal 9/97
Riverine 5/97
Riverine 9/97
0.004
0.007
0.005
0.011
0.00
0.03
0.00
0.02
0.017
0.064
0.016
0.032
0.001
0.010
0.115
0.143
0.54
0.71
0.20
0.55
0.72
0.32
0.39
0.54
1.9
1.7
1.8
2.5
21.3
20.2
14.7
15.0
TN:TP
4.7
21.9
10.5
TSI
30
36
31
35
PO4-P
TOTP
NH4-N
NO23-N
TKN
TN
Chl a, corr.
IN:IP
23
18.3
Trophic state indices were almost all “GOOD” (less than 50), with the exception of Stations 6 and
41 during September. For these stations, Secchi depths were low and increased the trophic state
index from “GOOD” to a “FAIR” designation. Spatial patterns of TSI, indicate higher values (4049) in the northern Mason Creek-Homosassa Bay area (Figure 8). Conductivity values, however,
were comparable between the northern and southern regions and there is not an obvious
dependence of TSI on salinity (Figure 9).
Figure 8.
Spatial distribution (by latitude) of trophic state indices; all samplings,
for Chassahowitzka riverine (RIVER) and coastal stations (COAST)
only.
Figure 9.
Lack of dependence of trophic state on salinity; coastal stations, all
samplings.
The higher TSI in the northern region imply that nutrient loadings may be greater in this area,
perhaps attributable to higher residential density or to influences of some of the larger watershed
rivers to the north (Withlacoochee, Suwannee Rivers). A riverine influence to the north is
consistent with the patterns of total and organic nitrogen increasing with salinity from the mouths
of the Chassahowitzka and the Homosassa Rivers (Figure 6).
Using non-parametric tests (Kruskal-Wallis), all nitrogen and phosphorus species, trophic state
indices, and inorganic N:P ratios were significantly (p < 0.05) higher in May than in September of
1997 for the coastal waters. The only significant changes for the riverine stations as a group were
increases in ortho-phosphorus and total nitrogen between May and September. Comparisons of
May 1996 with May 1997 detected significant changes (increases) only for salinity, temperature,
ammonia, and inorganic N:P ratios in the coastal waters and no changes for the riverine stations.
Secchi depths, with few exceptions were greater than bottom in the shallow coastal waters and so
changes in water clarity are untested.
Notable among the water quality data are the nitrate-nitrite-nitrogen concentrations collected at the
various spring vents and, in the Chassahowitzka and Homosassa main stems, the rapidity with
which it is removed from the water column. The most upstream sample on the Homosassa River
averaged 0.38 mg/l, even though not collected directly over a spring vent. Station CH#l and
POTTR averaged between 0.43 and 0.44 mg/l, the Chassahowitzka Main spring (RO) averaged
0.40 mg/l, while CRAB recorded slightly above 0.42 mg/l. Waters in the upper river were
strongly phosphorus-limited due to the excess of nitrate-nitrite-nitrogen present. Background
groundwater concentrations of unimpacted systems are estimated to be near 0.01 mg/l (Jones et
al., 1997). Figure 10 illustrates the removal of nitrate within a short distance downstream for the
Chassahowitzka stations.
Figure 10.
Gradient of nitrate-nitrite-nitrogen with respect to salinity; all
Chassahowitzka River samplings.
25
Most nitrate-nitrite-nitrogen had been consumed by Station 2.5, although in September,
measurable quantities were recorded at the mouth of the river. Perhaps in response to the
increased nutrient levels in September, riverine chlorophyll concentrations (Figure 11) and TSI
values (Figure 12) were also elevated with respect to the May samplings.
Figure 11.
Gradients of Chlorophyll a, corrected for pheophytin, with respect to
salinity; all Chassahowitzka River samplings.
Figure 12.
Gradients of Trophic State Index with respect to salinity; all
Chassahowitzka River samplings.
26
Gradients of organic nitrogen (Figure 13) are the reverse of the gradients typical of the estuarine
portions of surface watershed rivers. Even during September, when efforts were made to sample
the most offshore stations for nutrients and salinities were higher than previous samplings, no
decrease in nutrients with distance offshore was generally observed. The reverse gradient was
present in riverine stations, as well as in coastal locations and tidal creeks (see also Figure 6).
Export of organic nitrogen compounds from the forested wetlands and extensive marshes, in
addition to the influence of watershed rivers to the north, would contribute to the observed reverse
gradient.
Figure 13.
Reverse gradients of organic nitrogen with respect to salinity; all
Chassahowitzka samplings.
Interestingly, comparison of Chassahowitzka and Homosassa riverine gradients indicate that, for
a given fraction of saline waters, concentrations of nitrogen were higher in the Chassahowitzka
River (Figure 14). Patterns were similar for both May and September 1997. Sampling runs on
the two rivers were completed on successive days under similar tidal conditions and the results
indicate that for similar proportions of fresh and saline waters, the Chassahowitzka appears to have
a higher nitrogen load. Convergence of the offshore more saline stations indicate the influence
of a similar offshore water mass for each system. The smaller channel widths, fewer limestone
outcrops, and higher proportion of riverside marshes in the Chassahowitzka could all contribute
to this result. Despite the higher nitrogen concentrations in the Chassahowitzka, the Homosassa
River displayed higher chlorophyll (and therefore higher trophic state indices) for comparable
salinities (Figure 15). Residence times could differ sufficiently (higher in the Homosassa) to allow
phytoplankton to bloom in specific low salinity regions. Additionally, the organic nitrogen
exported from the wetlands and marshes as humic substances could be in a relatively nonbioavailable form.
27
Figure 14.
Comparison of total nitrogen as a function of salinity, in the
Chassahowitzka and Homosassa Rivers.
Phytoplankton
During 1997, phytoplankton and the dinoflagellate portions of the total community were manifestly
different from the 1996 sampling. From a fairly coherent and uniform ratio of
dinoflagellate:phytoplankton counts in 1996, dinoflagellates in 1997 samples formed a much
smaller fraction of the total population with a greater variation in the relationship.
28
Whereas dinoflagellates had formed 40-60% of the total organisms counted in May 1996,
dinoflagellates averaged 6% and 8 % of the total during the May and September samplings,
respectively (Figure 16). Total phytoplankton counts for the coastal areas averaged 84000,
187000, and 106000 organisms/l during the three samplings. For either single samplings or for
the combined database, neither salinity variations, nor a variety of nutrient concentrations
displayed a direct relationship to either total concentrations or dinoflagellate ratios.
Figure 16.
Dinoflagellate abundance as function of total phytoplankton concentration; riverine
and coastal stations. Scales truncated. A - May 1996, B - May 1997, C September 1997.
Biomass
A survey of wet weight biomass along a riverine transect in May and September 1997
demonstrated distinct spatial patterns in both algal and vascular abundance (Appendix D).
Although methods varied slightly between the two samplings, seasonal variations were also
captured. The May sampling recorded a maximum biomass of 12 kg/m2 at Station 1.3, distributed
evenly between rooted (Potamogeton pectinatus) and algal species. (May biomass values are the
arithmetic average of samples of “light”, “typical”, and “heavy” vegetation concentrations .) Total
biomass reached a minimum at Station 2.5 and 3, in an area of high salinity variation. Vascular
SAV (regardless of species) was primarily limited to the river (Figure 17), while algal species
(both attached and drift varieties) formed a bimodal pattern with a maximum in the low salinity
region of the river (generally filamentous green and blue-green varieties, minimal values at the
stations upstream of the mouth of the Chassahowitzka, and higher biomass values (Caulerpa
paspaloides, drift Rhodophyta) offshore.
29
Figure 17.
Riverine gradients of biomass, as the average of “light, ” “typical, ” and
“heavy” growth, during May 1997. Allocated into vascular submerged
aquatic vegetation (SAV) and macroalgal components.
During September, biomass samples represented the arithmetic average of five samples across the
river channel, or randomly placed in open waters. The bimodal pattern of algal biomass is still
present but the filamentous algae at the headspring was quite heavy, at 16 kg/mg2. Other. than this
station ‘biomass values were lower overall with vascular species forming a minima1 portion of the
total (Figure 18).
Figure 18.
Riverine gradients of biomass, as the average of five evenly spaced
(across channel) or randomly placed (open water) quadrats during
September 1997. Allocated into vascular submerged aquatic vegetation
(SAV) and macroalgal components.
30
Sediments
Results of sediment nutrient analyses are presented below and in Appendix E. For the purposes
of this exploratory sampling, total Kjeldahl nitrogen was assumed to be the majority of total
nitrogen in the sediments since nitrate porewater concentrations are typically low. (This
assumption will be tested in subsequent samplings.) During the May sample collection when
nutrients were analyzed, Stations R4 and 44 were on rocky substrate and no sample was possible.
Grain size results from both May and September samplings appear in Appendix E and reflect a
range of conditions within the river and coastal areas.
Within the upper river (Stations R0 through R1.3) water velocity is high in midchannel and
sediments consist of primarily the sand sized fraction. At Stations R1.7, R2, and at pockets
between limestone outcrops at Station 26 (and other offshore stations), sediments are very
unconsolidated and organic and nutrient content quite high, up to 10 mg TKN/kg dry sediment
(Figure 19).
Figure 19.
Riverine gradients of sedimentary total Kjeldahl nitrogen and total
phosphorus during May 1997. Rocky substrate and no sample at Stations
R4 and 44.
The unconsolidated nature and total nutrient content of sediment is mirrored by the spatial
distribution of sediment organic content (Figure 20). Interestingly, stations with high percent
organic content also have a high N:P (TKN:TOTP) ratio and could indicate that zones of organic
accumulation preferentially release phosphorus (under hypoxic or anoxic conditions), and that
organic sediments could act as a temporary nitrogen sink.
31
Figure 20.
Riverine gradients of sedimentary N:P ratios (N from TKN only) and
percent organics during May 1997.
The preferential mobilization of phosphorus from organic matter is supported by the spatial pattern
of phosphorus content per weight of organic matter in the sediment rather than per total dry weight
(Figure 21). In general, the phosphorus content of sedimentary organic material declines with
distance downstream, while nitrogen content although slightly variable, displays no monotonic
spatial trend and has a very direct relationship with sediment organic content (Figure 22). The
spatial pattern of phosphorus content of organic matter is also consistent with the strong
phosphorus limitation of upstream stations where nitrate-nitrite-nitrogen is present in quantity.
32
Figure 22.
Sediment total Kjeldahl nitrogen content as a function of percent organic
content during May 1997.
Summary
The oligotrophic nature of the nitrogen-limited coastal waters within the Chassahowitzka National
Wildlife Refuge are threatened by the prospect of increased nitrogen loads both from groundwater
discharges and from atmospheric deposition. Spring discharges of the Chassahowitzka and other
major spring complexes in the region already have nitrate-nitrite-nitrogen concentrations nearly
50 times that of pristine background conditions. Tidal variation in both discharge quality and
quantity are evidence of a complex hydrologic structure. Within the river systems, there are
numerous smaller spring vents which are largely unmapped, and there is also a high potential for
substantial quantities of submarine freshwater discharge. The relative amount of direct
groundwater discharge to the coastal waters is unknown at this time. Groundwaters discharging
to the region are approximately 10 to 50 years old and so even the most aggressive management
actions enacted at this time will only demonstrate nutrient reductions within a comparable time
span.
The sampling conducted for this report was designed to collect statistically robust baseline
information against which future samplings can be evaluated. Submerged aquatic vegetation (both
vascular and attached macroalgal) was identified to species level and abundances estimated at 30
coastal stations. Water quality samples were collected from 20 coastal stations, as well as along
the Chassahowitzka and Homosassa riverine salinity gradients. Samplings were conducted in May
and September 1997, during periods of both groundwater and rainfall deficits. Ancillary
samplings included both biomass and sediment nutrients along a gradient from upriver to offshore.
33
Initial work in the Chassahowitzka National Wildlife Refuge and Chassahowitzka River has
established that the coastal area supports a heterogeneous and stable community of submerged
aquatic vegetation. During the May samplings to date, vegetation in the coastal region was
dominated by a variety of drift algal species, while during all samplings, Caulerpa paspaloides
was the most frequently observed attached species. Declines in drift species frequency were noted
during September 1997, and Acetabularia was observed approximately half as often in 1997 as in
May 1996. Riverine biomass was bimodal with minimum values observed in the area of high
salinity variation. September biomass values were generally lower than May observations, and
the percentages of vascular plants in September data were very small.
Coastal waters are shallow and were well mixed during all samplings. September 1997 was
notable for higher salinities than had been observed previously. Spring discharges were also more
saline during this sampling event. Dissolved oxygen values were generally above instantaneous
criteria, but some violations ( < 4.0 mg/l) were recorded during early morning samplings. The
D.O. violations were typically in the northern portion of the Refuge coastal waters. September
D.O. levels and percent saturation values were significantly lower than during the prior two
samplings. Secchi depths were almost all greater than water column depths.
In coastal waters, inorganic nitrogen species, ortho-, and total phosphorus were generally quite
low. During May 1997, coastal waters were strongly nitrogen limited, while higher levels of
ammonium and nitrate-nitrite during September produced a more balanced water column. Trophic
state indices recorded two instances of the “FAIR” category during September, with the remaining
stations categorized as “GOOD”. Despite waters being more saline, all nitrogen and phosphorus
species, trophic state indices, and inorganic N:P ratios were significantly (Kruskal-Wallis) higher
in September as compared to May 1997.
Riverine water quality is distinguished by discharge of slightly saline waters from a complex of
spring vents at the head of the Chassahowitzka River. The nitrate-nitrite nitrogen concentration
is generally above 0.4 mg/l, but is rapidly removed from the water column by the extensive
vegetation in the freshwater portion of the river. Similar patterns of nitrate removal appear in the
Homosassa River, as well. Another interesting feature of riverine water quality in both systems
is that organic nitrogen is low at the spring vents and steadily increases with salinity, implying that
the nearshore oligotrophic waters are diluted with higher nutrient content, more saline waters,
perhaps representing nutrient loads from larger surface drainage rivers to the north. Export of
organic nitrogen from forested wetlands and marshes may also contribute to the reverse gradient.
In comparing the Chassahowitzka and Homosassa Rivers, the Chassahowitzka had more nitrogen
content for comparable salinity values. Conversely however, the Homosassa recorded higher
chlorophyll content and trophic state indices. The differences may be attributable to comparative
size and possible longer residence time of waters in the Homosassa, as well as the apparently
greater proportion of marshes along the Chassahowitzka.
34
Specific regions of the Chassahowitzka River and offshore areas have unconsolidated, highly
organic sediments. Sediment nutrient data indicate a preferential demobilization of phosphorus
from organics materials, with relatively constant proportions of nitrogen in organic matter.
Recommendations
Water column nutrients in the Refuge have been consistently low with many values (particularly
of inorganic nutrients) below analytical detection limits. The oligotrophic and nitrogen-limited
waters of the Refuge, however, are subjected to a variety of nutrient loadings. Inorganic nitrogen,
primarily nitrate-nitrite-nitrogen, is present in substantial quantities in many of the known spring
discharges, but is rapidly removed from the water column. Downstream loadings would be in the
form of dissolved and detrital compounds originating from the extensive biomass in the upper river.
In-Refuge sources could also include contributions by water exchange with the extensive coastal
marsh system. External sources would include the import from nearby major riverine systems, direct
contributions of groundwater to the coastal area, and atmospheric deposition of nitrate and
ammonium nitrogen. Direct atmospheric loading of nitrogen to Refuge waters is roughly comparable
to that from the major Chassahowitzka discharges, while the remaining sources are not quantifiable
at this time.
As known loadings of nitrogen are predicted to increase over time, monitoring of the Refuge water
column and biota should be designed to quantify areas and systems where changes can be predicted
or are expected. The primary goals should be the ability to detect temporal trends in water quality
and vegetation (particularly trends due to increasing nutrient loads), and to collect quantitative
information at intervals, focusing on information suitable for ecological modeling. To aid in trend
detections, water quality samplings would benefit from increased sampling frequency, while SAV
monitoring of a more stable community do not require the same frequency. Due to the complexity
of nitrogen sources to the Refuge itself, work within the Chassahowitzka River would simplify the
identification of direct relationships between tidal freshwater and brackish water community
responses and nitrogen loadings. Work within the coastal waters of the Refuge will allow the
detection of trends, but due to the numerous sources or potential sources of nitrogen, may not be
attributable to any given source.
35
LITERATURE CITED
Braun-Blanquet, J. 1932. Plant sociology: the study of plant communities. Transl. rev. and ed.
by C.D. Fuller and H.S. Conard. Hafner, London, 439 p.
Dawes, C.J. 1974. Marine algae of the west coast of Florida. Univ. Miami Press, 201 p.
Dixon, L.K. 1997. Data inventory, trend analysis, and recommended monitoring: Florida
Springs Coast. Mote Marine Laboratory Technical Report No. 536 to the Southwest Florida
Water Management District.
Earle, S. A. 1969. Phaeophyta of the eastern Gulf of Mexico. Phycologia 7: 71-254.
Estevez, E.D., L.K. Dixon and M.S. Flannery. 1991. West-coastal rivers of peninsular Florida,
Chapter 10 in R.J. Livingston (ed.), The Rivers of Florida. Springer-Verlag Ecological Studies
83. New York. 289 p.
Hand, J., V. Tauxe and M. Friedemann. 1988. 1988 Florida water quality assessment - 305(b)
technical appendix. Florida Dept. of Environ. Regul.
Hine, A.C. and D.F. Belknap. 1986. Recent geological history and modem sedimentary
processes of the Pasco, Hernando, and Citrus counties coastline. West-central Florida. Florida
Sea Grant Report 79. Gainesville, FL.
Iverson, R.L. and H.F. Bittaker. 1986. Seagrass distribution and abundance in eastern Gulf of
Mexico coastal waters. Est. Coastal Shelf Sci. 22: 577-602.
Jones, G.W., S.B. Upchurch, K.M. Champion and D.J. Dewitt. 1997. The water quality and
hydrology of Homosassa, Chassahowitzka, Weeki Wachee, and Aripeka spring complexe, Citrus
and Hernando Counties, Florida: Origin of increasing nitrate concentrations. Report prepared by
the Ambient Ground-Water Quality Monitoring Program, Southwest Florida Water Management
District.
Marshall, M.J. and J.R. Leverone. 1994. The distribution and effects of hypoxia on marine
organisms in Sarasota Bay. Mote Marine Laboratory Technical Report to the Sarasota Bay
National Estuary Program.
McNulty, J.K., W.N. Lindall, Jr. and J.E. Sykes. 1972. Cooperative Gulf of Mexico estuarine
inventory and study, Florida: Phase 1, area description. NOAA Technical Report NMFS
Circ-368. 126 p.
Officer, C.B. and J.H. Ryther. 1980. The possible importance of silicon in marine eutrophication.
Mar. Ecol. Prog. Ser. 62: 83-91.
36
Phillips, R.C. 1960. The ecology of marine plants at Crystal Bay, Florida. Quart. J. Fla. Acad.
Sci. 23: 328-337.
Southwest Florida Water Management District. 1994. SWFWMD Ambient Monitoring Program:
First biennial report of the ambient monitoring program including a report on water quality trends
in five central Florida springs. SWFWMD Resource Projects Department Environmental Section.
July.
Southwest Florida Water Management District. 1996. Hydrologic Conditions in the Southwest
Florida Water Management District, April 1996: with provisional updates through May 20.
Prepared by Hydrologic Data Section, Resources Data Department of SWFWMD.
Southwest Florida Water Management District. 1997a. Hydrologic Conditions in the Southwest
Florida Water Management District, May 1997. Prepared by Hydrologic Data Section, Resources
Data Department of SWFWMD.
Southwest Florida Water Management District. 1997b Hydrologic Conditions in the Southwest
Florida Water Management District, September 1997. Prepared by Hydrologic Data Section,
Resources Data Department of SWFWMD.
Wolfe, S.H. (ed.) 1990. An ecological characterization of the Florida Springs Coast:
Pithlachascotee to Waccasassa Rivers. U.S. Fish Wildl. Serv. Biol. Rep. 90(21). 323 p.
Yobbi, D.K. 1992. Effects of tidal stage and ground-water levels on the discharge and water quality
of springs in coastal Citrus and Hernando counties, Florida. U.S. Geol. Survey Water-Resources
Investigations Report 92-4069. 44 p.
37
Appendix A
SAV Quadrat Data
Station and Species Summary
Appendix B
In Situ Water Quality Data
Appendix C
Water Quality Data
Phytoplankton Data
Trophic State Indices
Appendix D
Biomass Data
Appendix E
Sediment Nutrient Data
Sediment Grain Size Data